Tumors exhibit altered uptake and utilization of nutrients, such as glucose and glutamine, to accommodate the tumor's need to accumulate biomass. In contrast to normally proliferative tissues, cells within a tumor are frequently starved for nutrients due to their high proliferation rate and unreliable vasculature. Therefore, the rewiring of cancer cell metabolism that occurs in response to nutrient limitation may present cancer specific vulnerabilities that can be the target of future anti-cancer therapies. Here, we propose to gain a better understanding of cancer metabolism by meeting the challenges of (1) determining how the tumor nutrient environment impacts cancer cell metabolism and (2) defining pathways which can be targeted as a consequence of this altered metabolism. In meeting these challenges, we will enable the fulfillment of our long-term goal: to characterize the metabolism of cancer in vivo and take advantage of the liabilities present due to this altered metabolism to identify essential genes which can be the target of future cancer therapies. We propose to address these challenges by two complementary Aims: (1) Determine those enzymes and pathways specifically essential to breast cancer cells in an orthotopic model of breast cancer and (2) using defined metabolic environments, determine those enzymes and pathways specifically essential to breast cancer cells in nutrient limited conditions. Completion of these first two Aims will give us the opportunity to (3) integrate the results from the two cancr cell systems and conduct targeted follow-up. Accomplishing the First Aim will require the implementation of an in vivo RNAi-based loss-of-function screen. This screen will be conducted using a pool of RNAi vectors targeting metabolic genes, enabling the construction of a pool of breast cancer cells, each of which exhibits suppression of a single enzyme. Upon in vivo or in vitro culture, the change in abundance of the RNAi construct will be measured by massively parallel DNA sequencing, and allow us to determine the essentiality of the gene which that construct suppresses. In the Second Aim we propose assessing the metabolite composition of murine or xenograft tumor models and patient tumor samples to identify key nutrients provided by the circulation that are depleted from individual tumors. Then, implementing a continuous medium replacement system that we have developed to grow cells in defined conditions where such key nutrients are limiting, we will define the adaptation to limiting key nutrients using a combination of expression profiling, metabolite profiling and RNAi-based screening, ultimately uncovering those enzymes or pathways essential for growth upon nutrient limitation. Finally, in the Third Aim we will have the opportunity to integrate the data from the first two Aims and identify metabolite, gene expression, or gene dependency profiles which are common or unique to the environments studied, with the goal of engaging in a targeted follow-up to gain a detailed mechanistic understanding of individual genes or pathways identified as essential in these environments.